nihui
/
ncnn
Not watched
1
0
Fork 0
Code
Releases 46 Wiki evaluate Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain HPC
You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.
3545 Commits
5 Branches
354 MB
1 Download
Tree: 7a0c19c856
Branches Tags
${ item.name }
Create branch ${ searchTerm }
from '7a0c19c856'
${ noResults }
ncnn/tests/testutil.h

testutil.h 5.5 kB
Raw Normal View History

use spdx license header, copyright Tencent (#6152)
11 months ago
​
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
​
​
​
SSE2 optimization pack (#2123) * SSE2: BatchNorm * Fixed batch norm in AVX configuration * Optimized register size switch * Attempt to pass CI * Attempt to pass CI * Bias op * Element wise ops * Support packing on x86 by default * Fixed macro range in bias * Use aligned read for packed data * Update testutil.h * Update pooling_x86.cpp * Support wasn SIMD * Fix emscripten compiler flags * fix build * more ci fix * concat x86 pack4 * flatten x86 pack4 * more x86 pack4 * ci pass * fix * enable sse2 mathfun * enable --experimental-wasm-simd Co-authored-by: nihui <shuizhuyuanluo@126.com> Co-authored-by: nihuini <nihuini@tencent.com>
5 years ago
format code style and setup restyled.io (#1840)
6 years ago
​
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
format code style and setup restyled.io (#1840)
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
innerproduct aarch64 use gemm (#2521) * perf(innerproduct-arm): add aarch64 gemm * fix(innerproduct): fix compilation errror * fix(armv7-innerproduct): fix armv7 compilation error * fix(innerproduct): fix gemm param * fix(int8): update mock scales and fix runtime error * fix(compilation): fix compilation error
5 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
architecture changes for int8 packing (#2771) * quantize and dequantize tests * unify activation and usability function * drop NCNN_REQUANT cmake option, test dequantize requantize pack8, fix webassembly build * benchmark use requantize int8 model
5 years ago
​
​
​
​
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
dynamic convolution weight (#3408)
4 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
adreno image shader + fp16 + fp16a (#1714) * wip * wip * fix * image and imageview can not be destroyed until command execution ends * fast copy path for tightly packed data * wip * texture load works * 1d 3d image * record clone image, multiple commands share one image reference * upload download image * layer forward accept vkimagemat * vkimagemat graph works * staging vkimagemat for passing dynamic parameters, macro for fp32+image shader, padding image shader * vkimagemat elemsize * convolution test pass * conv1x1s1 image shader * fast staging image allocator from host memory, pooling image shader * convolutiondepthwise image shader * innerproduct image shader * packing image shader * crop deconvolution image shader * resolve spirv binding types * image fp16 and fp16a, cast image shader * eltwise image shader * wip * absval image shader * deconvolutiondepthwise image shader * concat image shader, squeezenet works * noop split image shader * uniform precision hint * layer support_image_storage * wip * vulkan device utility operator * command is storage and packing option aware * fallback to cpu on image allocation failed, mobilenetssd works * flatten image shader, enable more test * ci test * check imgfp32 imgfp16 imgfp16a features * fix ci test * fix ci test * upgrade swiftshader * wip * opt aggressive * imgfp16p * opt none * convolution winograd image shader * fix flush range, fast copy path for continous buffer * minor fix * fix innerproduct * wip ... * wip * cast fix * packing test * wip * image fp16p is fp16p * wip * silence * more line info * code clean * softmax image shader
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
adreno image shader + fp16 + fp16a (#1714) * wip * wip * fix * image and imageview can not be destroyed until command execution ends * fast copy path for tightly packed data * wip * texture load works * 1d 3d image * record clone image, multiple commands share one image reference * upload download image * layer forward accept vkimagemat * vkimagemat graph works * staging vkimagemat for passing dynamic parameters, macro for fp32+image shader, padding image shader * vkimagemat elemsize * convolution test pass * conv1x1s1 image shader * fast staging image allocator from host memory, pooling image shader * convolutiondepthwise image shader * innerproduct image shader * packing image shader * crop deconvolution image shader * resolve spirv binding types * image fp16 and fp16a, cast image shader * eltwise image shader * wip * absval image shader * deconvolutiondepthwise image shader * concat image shader, squeezenet works * noop split image shader * uniform precision hint * layer support_image_storage * wip * vulkan device utility operator * command is storage and packing option aware * fallback to cpu on image allocation failed, mobilenetssd works * flatten image shader, enable more test * ci test * check imgfp32 imgfp16 imgfp16a features * fix ci test * fix ci test * upgrade swiftshader * wip * opt aggressive * imgfp16p * opt none * convolution winograd image shader * fix flush range, fast copy path for continous buffer * minor fix * fix innerproduct * wip ... * wip * cast fix * packing test * wip * image fp16p is fp16p * wip * silence * more line info * code clean * softmax image shader
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
adreno image shader + fp16 + fp16a (#1714) * wip * wip * fix * image and imageview can not be destroyed until command execution ends * fast copy path for tightly packed data * wip * texture load works * 1d 3d image * record clone image, multiple commands share one image reference * upload download image * layer forward accept vkimagemat * vkimagemat graph works * staging vkimagemat for passing dynamic parameters, macro for fp32+image shader, padding image shader * vkimagemat elemsize * convolution test pass * conv1x1s1 image shader * fast staging image allocator from host memory, pooling image shader * convolutiondepthwise image shader * innerproduct image shader * packing image shader * crop deconvolution image shader * resolve spirv binding types * image fp16 and fp16a, cast image shader * eltwise image shader * wip * absval image shader * deconvolutiondepthwise image shader * concat image shader, squeezenet works * noop split image shader * uniform precision hint * layer support_image_storage * wip * vulkan device utility operator * command is storage and packing option aware * fallback to cpu on image allocation failed, mobilenetssd works * flatten image shader, enable more test * ci test * check imgfp32 imgfp16 imgfp16a features * fix ci test * fix ci test * upgrade swiftshader * wip * opt aggressive * imgfp16p * opt none * convolution winograd image shader * fix flush range, fast copy path for continous buffer * minor fix * fix innerproduct * wip ... * wip * cast fix * packing test * wip * image fp16p is fp16p * wip * silence * more line info * code clean * softmax image shader
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
adreno image shader + fp16 + fp16a (#1714) * wip * wip * fix * image and imageview can not be destroyed until command execution ends * fast copy path for tightly packed data * wip * texture load works * 1d 3d image * record clone image, multiple commands share one image reference * upload download image * layer forward accept vkimagemat * vkimagemat graph works * staging vkimagemat for passing dynamic parameters, macro for fp32+image shader, padding image shader * vkimagemat elemsize * convolution test pass * conv1x1s1 image shader * fast staging image allocator from host memory, pooling image shader * convolutiondepthwise image shader * innerproduct image shader * packing image shader * crop deconvolution image shader * resolve spirv binding types * image fp16 and fp16a, cast image shader * eltwise image shader * wip * absval image shader * deconvolutiondepthwise image shader * concat image shader, squeezenet works * noop split image shader * uniform precision hint * layer support_image_storage * wip * vulkan device utility operator * command is storage and packing option aware * fallback to cpu on image allocation failed, mobilenetssd works * flatten image shader, enable more test * ci test * check imgfp32 imgfp16 imgfp16a features * fix ci test * fix ci test * upgrade swiftshader * wip * opt aggressive * imgfp16p * opt none * convolution winograd image shader * fix flush range, fast copy path for continous buffer * minor fix * fix innerproduct * wip ... * wip * cast fix * packing test * wip * image fp16p is fp16p * wip * silence * more line info * code clean * softmax image shader
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
​
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
​
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
​
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
armv8.2 infrastructure (#1856) * runtime cpu dispatch * force thread one * disable openmp for coverage * simplify test layer * print NCNN_TARGET_ARCH * less ci build variants * weight fp16 storage option * test convdw int8 * apple a12 a13 * ncnn_add_layer ncnn_add_shader cmake macro
5 years ago
shader shape specialization constant and basic local group size partition (#1523) * use Mat class for Shape description * shape specialization constant in compute shader * wip * wip * test forward_inplace, add binaryop unaryop sigmoid test * fix arm unaryop test * fix arm binaryop test * make shape hint optional, cast int8 to fp32, add cast test * wip * follow the good and old local size setting for conv1x1 * the optimal local size rewrite * fix build on msvc * add permute shader for all packing layout, add permute test * concat and slice patial shape constant, slice test * fix slice test * interp test * add lrn test, test packing layout implicitly * add eltwise test * add normalize test * add instancenorm test * reorg shape constant * simple local group size partition * add shape constant param
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
test custom option, update ci (#4609) * early return for cpu test * make nvidia driver happy * fix gemm x86 threading
3 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
test custom option, update ci (#4609) * early return for cpu test * make nvidia driver happy * fix gemm x86 threading
3 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
test custom option, update ci (#4609) * early return for cpu test * make nvidia driver happy * fix gemm x86 threading
3 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
shader shape specialization constant and basic local group size partition (#1523) * use Mat class for Shape description * shape specialization constant in compute shader * wip * wip * test forward_inplace, add binaryop unaryop sigmoid test * fix arm unaryop test * fix arm binaryop test * make shape hint optional, cast int8 to fp32, add cast test * wip * follow the good and old local size setting for conv1x1 * the optimal local size rewrite * fix build on msvc * add permute shader for all packing layout, add permute test * concat and slice patial shape constant, slice test * fix slice test * interp test * add lrn test, test packing layout implicitly * add eltwise test * add normalize test * add instancenorm test * reorg shape constant * simple local group size partition * add shape constant param
6 years ago
create layer decoupled (#5258) * create layer decoupled * no more virtual public * allow build test with shared library * decouple cpu vulkan * drop old scripts
2 years ago
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
test x86 arm convolution oom (#5492) * skip mips loongarch riscv oom test atm * test softmax oom
2 years ago
​
​
​
​
​
​
​
​
​
the initial auto test (#1464) * cpu test * wip * ci run test * travis ci for arm64 * arm64 ctest * copy vulkan loader * wip * run * Update ccpp.yml * gpu test * swiftshader * cache macos swiftshader * try MoltenVK * try vulkaninfo * give swiftshader another try * disable failed macos gpu test * more conv test, fix conv3x3s1 gpu test fail * fix deconvolution test * dilation test * cmake option to build tests * ncnn_add_layer_test macro * host barrier before upload and after download, handle packing layout option * test packing layout * wip * wip * merge deconvolution packing and non-packing code * merge convolution packing and non-packing code * pass top_blob_count param * fix build * take care of non-coherent mappable memory
6 years ago
 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108 
  1. // Copyright 2019 Tencent

  2. // SPDX-License-Identifier: BSD-3-Clause

  3. 

  4. #ifndef TESTUTIL_H

  5. #define TESTUTIL_H

  6. 

  7. #include "cpu.h"

  8. #include "layer.h"

  9. #include "mat.h"

  10. 

  11. #include <stdio.h>

  12. #include <stdint.h>

  13. #include <stdlib.h>

  14. 

  15. #define TEST_LAYER_DISABLE_AUTO_INPUT_PACKING (1 << 0)

  16. #define TEST_LAYER_DISABLE_AUTO_INPUT_CASTING (1 << 1)

  17. #define TEST_LAYER_DISABLE_GPU_TESTING        (1 << 2)

  18. #define TEST_LAYER_ENABLE_FORCE_INPUT_PACK8   (1 << 3)

  19. 

  20. void SRAND(int seed);

  21. 

  22. uint64_t RAND();

  23. 

  24. float RandomFloat(float a = -1.2f, float b = 1.2f);

  25. 

  26. int RandomInt(int a = -10000, int b = 10000);

  27. 

  28. signed char RandomS8();

  29. 

  30. void Randomize(ncnn::Mat& m, float a = -1.2f, float b = 1.2f);

  31. 

  32. void RandomizeInt(ncnn::Mat& m, int a = -10000, int b = 10000);

  33. 

  34. void RandomizeS8(ncnn::Mat& m);

  35. 

  36. ncnn::Mat RandomMat(int w, float a = -1.2f, float b = 1.2f);

  37. 

  38. ncnn::Mat RandomMat(int w, int h, float a = -1.2f, float b = 1.2f);

  39. 

  40. ncnn::Mat RandomMat(int w, int h, int c, float a = -1.2f, float b = 1.2f);

  41. 

  42. ncnn::Mat RandomMat(int w, int h, int d, int c, float a = -1.2f, float b = 1.2f);

  43. 

  44. ncnn::Mat RandomIntMat(int w);

  45. 

  46. ncnn::Mat RandomIntMat(int w, int h);

  47. 

  48. ncnn::Mat RandomIntMat(int w, int h, int c);

  49. 

  50. ncnn::Mat RandomIntMat(int w, int h, int d, int c);

  51. 

  52. ncnn::Mat RandomS8Mat(int w);

  53. 

  54. ncnn::Mat RandomS8Mat(int w, int h);

  55. 

  56. ncnn::Mat RandomS8Mat(int w, int h, int c);

  57. 

  58. ncnn::Mat RandomS8Mat(int w, int h, int d, int c);

  59. 

  60. ncnn::Mat scales_mat(const ncnn::Mat& mat, int m, int k, int ldx);

  61. 

  62. bool NearlyEqual(float a, float b, float epsilon);

  63. 

  64. int Compare(const ncnn::Mat& a, const ncnn::Mat& b, float epsilon = 0.001);

  65. 

  66. int CompareMat(const ncnn::Mat& a, const ncnn::Mat& b, float epsilon = 0.001);

  67. 

  68. int CompareMat(const std::vector<ncnn::Mat>& a, const std::vector<ncnn::Mat>& b, float epsilon = 0.001);

  69. 

  70. int test_layer_naive(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const std::vector<ncnn::Mat>& a, int top_blob_count, std::vector<ncnn::Mat>& b, void (*func)(ncnn::Layer*), int flag);

  71. 

  72. int test_layer_cpu(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const std::vector<ncnn::Mat>& a, int top_blob_count, std::vector<ncnn::Mat>& c, const std::vector<ncnn::Mat>& top_shapes, void (*func)(ncnn::Layer*), int flag);

  73. 

  74. #if NCNN_VULKAN

  75. int test_layer_gpu(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const std::vector<ncnn::Mat>& a, int top_blob_count, std::vector<ncnn::Mat>& d, const std::vector<ncnn::Mat>& top_shapes, void (*func)(ncnn::Layer*), int flag);

  76. #endif // NCNN_VULKAN

  77. 

  78. int test_layer(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const std::vector<ncnn::Mat>& a, int top_blob_count, const std::vector<ncnn::Mat>& top_shapes = std::vector<ncnn::Mat>(), float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  79. 

  80. int test_layer_naive(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Mat& a, ncnn::Mat& b, void (*func)(ncnn::Layer*), int flag);

  81. 

  82. int test_layer_cpu(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const ncnn::Mat& a, ncnn::Mat& c, const ncnn::Mat& top_shape, void (*func)(ncnn::Layer*), int flag);

  83. 

  84. #if NCNN_VULKAN

  85. int test_layer_gpu(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const ncnn::Mat& a, ncnn::Mat& d, const ncnn::Mat& top_shape, void (*func)(ncnn::Layer*), int flag);

  86. #endif // NCNN_VULKAN

  87. 

  88. int test_layer(int typeindex, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& _opt, const ncnn::Mat& a, const ncnn::Mat& top_shape = ncnn::Mat(), float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  89. 

  90. int test_layer_opt(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& opt, const std::vector<ncnn::Mat>& a, int top_blob_count = 1, float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  91. 

  92. int test_layer_opt(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& opt, const ncnn::Mat& a, float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  93. 

  94. int test_layer(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const std::vector<ncnn::Mat>& a, int top_blob_count = 1, float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  95. 

  96. int test_layer(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Mat& a, float epsilon = 0.001, void (*func)(ncnn::Layer*) = 0, int flag = 0);

  97. 

  98. // oom test

  99. 

  100. int test_layer_oom_opt(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& opt, const std::vector<ncnn::Mat>& a, int top_blob_count = 1, int flag = 0);

  101. 

  102. int test_layer_oom_opt(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Option& opt, const ncnn::Mat& a, int flag = 0);

  103. 

  104. int test_layer_oom(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const std::vector<ncnn::Mat>& a, int top_blob_count = 1, int flag = 0);

  105. 

  106. int test_layer_oom(const char* layer_type, const ncnn::ParamDict& pd, const std::vector<ncnn::Mat>& weights, const ncnn::Mat& a, int flag = 0);

  107. 

  108. #endif // TESTUTIL_H

No Description